Lateral Connections in Denoising Autoencoders Support Supervised Learning

Abstract

We show how a deep denoising autoencoder with lateral connections can be used as an auxiliary unsupervised learning task to support supervised learning. The proposed model is trained to minimize simultaneously the sum of supervised and unsupervised cost functions by back-propagation, avoiding the need for layer-wise pretraining. It improves the state of the art significantly in the permutation-invariant MNIST classification task.

References in corpus (1)

• Denoising autoencoder with modulated lateral connections learns invariant representations of natural images

Cited by in corpus (5)

• Semi-Supervised Learning with Ladder Networks
• Distributional Smoothing with Virtual Adversarial Training
• Stacked What-Where Auto-encoders
• Semi-Supervised Speech Emotion Recognition with Ladder Networks
• Ladder Networks for Emotion Recognition: Using Unsupervised Auxiliary Tasks to Improve Predictions of Emotional Attributes